Abstract
This paper investigates a general framework tor learning concepts that allows to generate accurate and comprehensible concept representations. It is known that biases used in learning algorithms directly affect their performance as well as their comprehensibility. A critical problem is that, most of the time, the most “comprehensible” representations are not the best performer in terms of classification! In this paper, we argue that concept learning systems should employ Multiple-Knowledge Representation: a deep knowledge level optimised from recognition (classification task) and a shallow one optimised for comprehensibility (description task). Such a model of concept learning assumes that the system can use an interpretation function of the deep knowledge level to build an approximately correct comprehensible description of it. This approach is illustrated through our GEM system which learns concepts in a numerical attribute space using a Neural Network representation as the deep knowledge level and symbolic rules as the shallow level.
Chapter PDF
Similar content being viewed by others
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
References
Aha W. David, Kibler D., Albert K. M. (1991) Instance-Based Learning Algorithms, Machine Learning vol.6, n° 1, January 1991, Kluwer Academic Publishers.
Bergadano F., Esposito F., Rouveirol C. and Wrobel S. (1991) Evaluating and Changing Representation in Concept Acquisition, Proceedings of the European Working Session on Learning, Springer Verlag
Bergadano F., Matwin S., Michalski R.S., Zhang J. (1992) Learning Two-Tiered Descriptions of Flexible Concepts: The Poseidon System, Machine Learning vol.8, n° 1, Kluwer Academic Publishers.
Benjamin Paul D. (Ed) (1990) Change of Representation and Inductive Bias, Kluwer Academic Publishers.
Brodley Carla E. (1993) Addressing the Selective Superiority Problem: Automatic Algorithm/ Model Class Selection, Proceedings of the Tenth International Conference on Machine Learning ML'93, Morgan Kaufmann.
Buntine Wray (1989) Learning Classification Rules using Bayes, Proceedings of the Sixth International Workshop on Machine Learning ML'89. Morgan Kaufmann.
Clark P. and Niblett T. (1989) The CN2 Induction Algorithm, Machine Learning Vol.3 n°4, March 1989, Kluwer Academic Publishers.
Decaestecker C. (1993) NNP: a neural net classifier using prototypes, Proceedings of the IEEE International Conference on Neural Networks, San Fransisco.
Esposito Floriana, Malerba Donato and Semeraro Giovanni (1991) Flexible Matching for Noisy Structural Descriptions, Proceedings of the Thirteenth International Joint Conference on Artificial Intelligence IJCAI'91, Morgan Kaufmann.
Goodman R.M., Higgins C.M. & Miller J.W. (1992) Rule-based neural networks for classification and probability estimation, Neural Computation Vol.4 n° 6.
Hertz J., Krogh A. & Palmer R.G. (1991) Introduction to the theory of neural computation, Addison-Wesley.
Knight K. and Gil Y. (1991), Automated Rationalization, Proceedings of the First International Worshop on Multistrategy Learning, Ed. by R.S. Michalski and G. Tecuci, Center of Artificial Intelligence, George Mason University.
Kohonen T. (1990) The Self-Organizing Map, Proceedings of the IEEE, vol. 78, N°9.
Iba W., Wogulis J., Langley P. (1988) Trading Off Simplicity and Coverage in Incremental Concept Learning, Proceedings of the Fith International Conference on Machine Learning ML'88, Morgan Kaufman.
Michalski Ryszard S. (1983) A Theory and Methodology of Inductive Learning. Machine Learning, An Artificial Intelligence Approach. Ed. by Ryszard S. Michalski, Jaime G. Carbonell and Tom M. Mitchell, Tioga Publishing.
Michalski Ryszard S. (1990) Learning Flexible Concepts: Fundamental Ideas and Method Based on Two-Tiered Representation, Machine Learning: An Artificial Intelligence Approach Vol. III. Edited by Y. Kodratoff and Ryszard S. Michalski, Morgan Kaufmann.
Quinlan J. Ross (1986a) Induction of Decision Trees. Machine Learning Vol 1, n°1, Kluwer Academic Publishers.
Salzberg Steven (1991), A Nearest Hyperrectangle Learning Method. Machine Learning vol. 6, n° 3, May 1991, Kluwer Academic Publishers.
Samkar A. & Mammone R.J. (1991) Neural Tree Networks. Neural Networks, Therory and Applications. R.J. Mammone & Y. Zeevi Eds, Academic Press.
Stepp Robert E. and Michalski Ryszard S. (1983) Conceptual Clustering: Inventing Goal-oriented Classification of Structured Objects, Machine Learning, An Artificial Intelligence Approach volII. Ed. by Ryszard S. Michalski, Jaime G. Carbonell and Tom M. Mitchell, Morgan Kaufmann.
Towell G. G. and Shavlik J. (1993) Extracting Refined Rules from Knowledge-Based Neural Networks, Machine Learning, vol. 13, n° 1, Kluwer Academic Publishers.
Tschichold N., Ghazvini M. and Diez D. (1992), M-RCE: a self configuring ANN with rule extraction capabilities, Proceedings of the International Conference on Artificial Neural Networks ICANN'92, Brighton.
Utgoff Paul E. (1986) Machine Learning of Inductive Bias. Kluwer Academic Publishers.
Utgoff P.E. (1988) Perceptron Trees: A case Study in Hybrid Concepts Representations, Proceedings of AAAI-88.
Van de Merckt T. (1992) NFDT: A Sytem that Learns Flexible Concepts based on Decision Trees for Numerical Attributes. Proceedings of the Ninth International Conference on Machine Learning ML'92, Morgan Kaufmann.
Van de Merckt T. and Decaestecker C. (1994), An unifying framework for analysing bias in Similarity Based Learning, Proceedings of the MlNet Workshop on Declarative Bias, European Conference on Machine Learning, Catania.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1995 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Van de Merckt, T., Decaestecker, C. (1995). Multiple-Knowledge Representations in concept learning. In: Lavrac, N., Wrobel, S. (eds) Machine Learning: ECML-95. ECML 1995. Lecture Notes in Computer Science, vol 912. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-59286-5_59
Download citation
DOI: https://doi.org/10.1007/3-540-59286-5_59
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-59286-0
Online ISBN: 978-3-540-49232-0
eBook Packages: Springer Book Archive